← Back

RE-LY

Dabigatran versus Warfarin in Patients with Atrial Fibrillation

Share Share Copied! Compare

Year of Publication: 2009

Authors: Stuart J. Connolly, M.D., Michael D. Ezekowitz, ..., Lars Wallentin

Journal: The New England Journal of Medicine

Citation: N Engl J Med 2009;361:1139-51.

Link: https://doi.org/10.1056/NEJMoa0905561

PDF: https://www.nejm.org/doi/pdf/10.1056/NEJMoa0905561

Clinical Question

In patients with atrial fibrillation at risk for stroke, are fixed doses of dabigatran (110 mg or 150 mg twice daily) noninferior to adjusted-dose warfarin for preventing the primary outcome of stroke or systemic embolism?

Study Overview

Objective

To compare dabigatran (110 mg or 150 mg BID) with warfarin for stroke prevention in atrial fibrillation.

Study Summary

Dabigatran 150 mg was superior to warfarin for stroke/systemic embolism prevention in non-valvular AF;
Dabigatran 110 mg was noninferior to warfarin with less major bleeding.

Intervention

Fixed-dose dabigatran (110 or 150 mg BID) vs. adjusted-dose warfarin (INR 2–3). Median follow-up: 2 years.

Patients per Arm

110 mg: ~6015, 150 mg: ~6076, Warfarin: ~6022

Bottom Line

In patients with atrial fibrillation, dabigatran 150 mg twice daily was superior to warfarin for the prevention of stroke or systemic embolism, with similar rates of major bleeding but lower rates of life-threatening and intracranial bleeding. The 110 mg dose was noninferior to warfarin for stroke prevention and was associated with significantly lower rates of major bleeding.

        Major Points
        First major DOAC trial — RE-LY launched the era of direct oral anticoagulants by demonstrating that dabigatran (a direct thrombin inhibitor) could match or beat warfarin for AF stroke prevention.
18,113 patients across 951 centers in 44 countries. Unique three-arm design: two blinded dabigatran doses (110 mg and 150 mg BID) vs open-label warfarin (PROBE design for warfarin comparison).
Dabigatran 150 mg BID was SUPERIOR to warfarin: stroke/systemic embolism 1.11% vs 1.69%/yr (RR 0.66, 95% CI 0.53–0.82, p<0.001 for superiority). The only DOAC to demonstrate superiority over warfarin for stroke prevention.
Dabigatran 110 mg BID was NONINFERIOR to warfarin: 1.53% vs 1.69%/yr (RR 0.91, p<0.001 for noninferiority). This dose was approved outside the US but not in the US (FDA approved only 150 mg and 75 mg).
Both doses dramatically reduced hemorrhagic stroke vs warfarin: 150 mg: 0.10% vs 0.38%/yr (RR 0.26, p<0.001); 110 mg: 0.12% vs 0.38%/yr (RR 0.31, p<0.001). This was the most consistent benefit across all DOACs vs warfarin.
Major bleeding: 150 mg was similar to warfarin (3.11% vs 3.36%/yr, p=0.31); 110 mg was significantly lower (2.71%/yr, p=0.003). GI bleeding was higher with 150 mg (1.51% vs 1.02%/yr, p<0.001).
MI signal: both dabigatran doses showed numerically higher MI rates (150 mg: 0.74% vs 0.53%/yr, RR 1.38, p=0.048) — a controversial finding possibly explained by warfarin's anti-thrombotic properties on coronary plaque.
Dyspepsia was significantly more common with dabigatran (11.3% vs 5.8%) — the tartaric acid core (required for low-pH absorption) may partly explain the increased dyspeptic symptoms, and it was a major driver of drug discontinuation.
Idarucizumab (Praxbind) was subsequently developed as a specific reversal agent for dabigatran — the first DOAC reversal agent approved.
Led to FDA approval of dabigatran in 2010, making it the first DOAC available in the US. Launched a paradigm shift from warfarin to DOACs for AF, followed by ROCKET AF (rivaroxaban), ARISTOTLE (apixaban), and ENGAGE AF-TIMI 48 (edoxaban).

      

Design

Study Type: Randomized, noninferiority trial.

Randomization: 1

Blinding: Dabigatran doses were administered in a blinded fashion; warfarin was administered in an open-label fashion. Outcomes were adjudicated by a blinded committee.

Enrollment Period: December 22, 2005, to December 15, 2007.

Follow-up Duration: Median of 2.0 years.

Centers: 951

Countries: 44 countries

Sample Size: 18113

Analysis: Intention-to-treat.

Inclusion Criteria

Atrial fibrillation documented on ECG at screening or within 6 months before screening.
At least one stroke risk factor: previous stroke or TIA, LVEF <40%, NYHA class II or higher heart failure symptoms within 6 months, age ≥75 years, or age 65–74 years plus diabetes mellitus, hypertension, or coronary artery disease.
Paroxysmal, persistent, or permanent AF accepted.

Exclusion Criteria

Severe heart-valve disorder.
Stroke within 14 days or severe stroke within 6 months.
Condition that increased the risk of hemorrhage.
Creatinine clearance <30 ml per minute.
Active liver disease.
Pregnancy.

Baseline Characteristics

Characteristic	Control	Active
Age-yr	71.6±8.6	71.5±8.8
Male sex no./total no. (%)	3809/6022 (63.3)	3840/6076 (63.2)
CHADS2 score	2.1±1.1	2.2±1.2
Previous stroke or transient ischemic attack-no./total no. (%)	1195/6022 (19.8)	1233/6076 (20.3)
Prior myocardial infarction no./total no. (%)	968/6022 (16.1)	1029/6076 (16.9)
Heart failure no./total no. (%)	1922/6022 (31.9)	1934/6076 (31.8)
Diabetes mellitus no./total no. (%)	1410/6022 (23.4)	1402/6076 (23.1)
Hypertension-no./total no. (%)	4750/6022 (78.9)	4795/6076 (78.9)
Aspirin use at baseline - no./total no. (%)	2442/6017 (40.6)	2352/6075 (38.7)

Arms

Field	Control	Dabigatran 110 mg	Dabigatran 150 mg
Intervention	Adjusted-dose warfarin targeting INR 2.0–3.0, administered open-label with local INR monitoring at least monthly. Mean time in therapeutic range (TTR) was 64% overall — moderate quality anticoagulation control. Dose adjustments per local practice. Random assignment was made after consent; the operations committee later enforced balanced enrollment between patients with prior long-term VKA therapy (~50%) and VKA-naive patients (no run-in period was used).	Dabigatran etexilate 110 mg BID (lower dose). Direct thrombin (Factor IIa) inhibitor — the first oral direct thrombin inhibitor for AF. Fixed dose, no monitoring required. Blinded against 150 mg dose. Tartaric acid core for absorption (may contribute to dyspepsia). Renal clearance ~80% — dose depends on renal function. Not approved in US; approved in Europe/Canada/Japan for patients at higher bleeding risk or age ≥80.	Dabigatran etexilate 150 mg BID (higher dose). Same mechanism as 110 mg. Blinded against 110 mg dose. This dose achieved superiority over warfarin for efficacy. FDA-approved dose in the US for CrCl >30 mL/min (US also approved 75 mg for CrCl 15–30 mL/min, based on PK modeling, not tested in RE-LY). Idarucizumab (Praxbind) is the specific reversal agent.
Duration	Median 2.0 years	Median 2.0 years	Median 2.0 years

Outcomes

Outcome	Type	Control	Intervention	HR / OR / RR	P-value
The primary outcome was stroke or systemic embolism. Effect measure reported by the paper is relative risk (not hazard ratio).	Primary	1.69% per year	1.11% per year (for 150 mg dose)	0.66	<0.001 for superiority
Hemorrhagic stroke	Secondary	0.38% per year	0.10% per year (for 150 mg dose)	RR 0.26 (95% CI, 0.14 to 0.49)	<0.001
Ischemic or unspecified stroke	Secondary	1.20% per year	0.92% per year (for 150 mg dose)	RR 0.76 (95% CI, 0.60 to 0.98)	0.03
Death from any cause	Secondary	4.13% per year	3.64% per year (for 150 mg dose)	RR 0.88 (95% CI, 0.77 to 1.00)	0.051
Myocardial infarction	Secondary	0.53% per year	0.74% per year (for 150 mg dose)	RR 1.38 (95% CI, 1.00 to 1.91)	0.048
Major bleeding	Adverse	3.36% per year	3.11% per year (for 150 mg dose)	RR 0.93 (95% CI, 0.81 to 1.07)	0.31
Intracranial bleeding	Adverse	0.74% per year	0.30% per year (for 150 mg dose)	RR 0.40 (95% CI, 0.27-0.60)	<0.001
Major gastrointestinal bleeding	Adverse	1.02% per year	1.51% per year (for 150 mg dose)	RR 1.50 (95% CI, 1.19-1.89)	<0.001
Dyspepsia	Adverse	5.8%	11.3% (for 150 mg dose)		<0.001

Subgroup Analysis

Pre-specified Figure 2 subgroups (dabigatran 150 mg and 110 mg vs warfarin for stroke or systemic embolism): long-term VKA therapy, sex, body-mass index, weight, ethnic group, creatinine clearance, CHADS2 score (0 or 1 / 2 / ≥3), symptomatic heart failure, hypertension, diabetes, prior stroke or TIA, geographic region, baseline aspirin use, amiodarone use, and PPI use. No significant treatment-by-subgroup interaction was seen for dabigatran at either dose across these subgroups. In particular, despite ~80% renal excretion, no significant interaction was noted across levels of baseline creatinine clearance, and no significant interaction was seen with prior long-term VKA therapy.

Criticisms

PROBE design with open-label warfarin — the most criticized aspect of the trial. Knowledge of treatment assignment could influence anticoagulation management and event reporting, despite blinded adjudication.
MI signal with both dabigatran doses (150 mg: RR 1.38, p=0.048) was concerning and unexpected — meta-analyses later suggested a ~30% increased MI risk, potentially limiting use in patients with coronary disease.
GI bleeding was higher with 150 mg (1.51% vs 1.02%/yr, p<0.001) — particularly in elderly patients, making dose selection complex.
Dyspepsia rate of 11.3% with dabigatran (vs 5.8% warfarin) — the low-pH tartaric acid formulation required for absorption may partly explain the increased dyspeptic symptoms and drug discontinuation in real-world use.
Warfarin TTR of 64% was moderate — raising questions about whether the trial demonstrated dabigatran superiority or warfarin inferiority in centers with lower-quality INR control.
No dose reduction criteria were built into the trial — the FDA later approved 75 mg for CrCl 15–30 mL/min based on PK modeling alone, not RE-LY data.
Industry-sponsored (Boehringer Ingelheim) — and there were later controversies about bleeding event reporting and FDA inspection findings.
Drug interactions not well characterized — dabigatran is a P-glycoprotein substrate, and concurrent P-gp inhibitors (amiodarone, dronedarone, verapamil) increase levels significantly.
No head-to-head comparison with other DOACs — clinical decision-making between dabigatran, rivaroxaban, apixaban, and edoxaban relies on indirect comparisons.

Funding

Boehringer Ingelheim.

Based on: RE-LY (The New England Journal of Medicine, 2009)